Bridged Polycyclic Compounds. V. The Addition of p-Thiocresol to

Feb. 5 , 1958

,ADDITION OF

lCONTRIBUrION FROM T H E

p-THIOCRESOL TO NORBORNADIEKE

DEPARTMEXT O F CHEMISTRY, UXIVERSITY

635

O F COLORADO]

Bridged Polycyclic Compounds. V. The Addition of p-Thiocresol to Norbornadiene. The Question of Non-classical Free Radicals1,2 BY STANLEY J. CRISTOL, GORDOX D. BRINDELL AKD

JAMES A.

REEDER

RECEIVED JULY 29, 1957 Addition of thiopheilol or p-thiocresol to norburnadiene ( I ) under free-radical conditions leads to mixtures of exo-hiorborneii-2-yl aryl tliioethers via 1,a-additioii and 3-nortricyclyl aryl thioethers via homoconjugative addition, Structureproof work on the products is described. Variations of the product composition with concentration of the mercaptan in the reaction mixture demonstrate t h a t the reaction involves addition of the thiol radical to the olefin t o give an exoarylthionorbornenyl radical which may ( a ) accept a hydrogen atom from a mercaptan molecule t o give the 1.2-addition product or (b) isomerize t o a n arylthionortricyclyl radical, \+ hich transfers to give the homoconjugative addition product. Conclusive evidence is given to indicate that the two addition products d o not result from a common intermediate such as the “non-classical” homoallglic radical YIII.

The addition of mercaptans to olefins has been shown t o proceed aia a chain mechanism involving free-radical intermediates, 3 , 4 viz. RS’

I I

1 1

KSC-C*

+ >C=C
. L. Sims, ibid.. 77. 3465 (11333); (c) H. L Goering, D I. K e l y r a and D. R- Larsen, ibid.,7 8 , 348 (1556). (8) (a) P. S.Skell a n d R . C. Woodworth, zbid., 7 7 , 4638 (105.5); ( b ) P.S.Skcll, R C. n’oorlwortli a n d J . €1. L I c S u i n a r a , i b d , 79, 1233 (1927). (9) (a) I;. G . Bordwell and I\’. A. H e w e t t , Abstracts u f t h e I2titli Meeting of t h e American Chemical Society, New York, K. Y . , S e p tember. 1934, p. 6 - 0 ; (b) F. G . Bordwell a n d S . P. Seiireiter, 4 b -

stracts of t h e 131st Meeting of t h e American Chemical Society, Miami, Fla., April, 1557, p . 1 0 - 0 . (10) J . D. Roberts, E R. Truinbull. J r . , It’. Bennett and I.lfter two distillations. Ratio of rearranged to unrearranged product, assuming t h a t isolated product represents whole product. Ratio of rearranged to uiirearrnngcd prodnct, aswming that all losses r e r c olefin ITr. 1: 1 inolar ratio of 111 and .IrSH, no solvent. f 1: 1 molar ratio of 111 and ArSH, diluted with 10 volumes of chlorobenzene. 0 ArSH added dropwise to an equimolar quantity of I11 in 10 \-olumes of chlorobenzene. Same as expt. 2, but heated to 10.5'. Same as expt. 1, but with 10 volumes of ethylene glycol to act as heat absorber. i Xverage of 5 r1crermin:itions. k .iverage of 3 determinations.

le 2.'

80.2 84.8

Since the addition reaction is highly exothermic. the reaction mixture becomes hot when no diluent is used. It seemed possible that the increase in rearrangement with dilution of the reaction mixture might reflect a temperature effect rather thmi a concentration effect. Thus in the fourth experiment the diluted reactants were mixed a t 105", and in the fifth experiment the undiluted reactxnts

were stirred with ethylene glycol, which did not dissolve the reactants but was present to act as a coolant. Comparison of experiments 4 and 2 shows that if there is any significant temperature effect it is small and not in the right direction to cause the dilution effect. A comparison of experiments 1 and 5 again indicates the unimportance of the temperature effect. Thus the experimental results demonstrate quite clearly that the mesomeric radical intermediate VI11 cannot be the cominoii progenitor of IV and V j but that two separate radical species must be involved. For the present, there seems t o be no experimental reason to assume that these radicals are not the classical free radicals VI and 1711.27

It was observed that a p-thiocresoxy radical reacts more rapidly with norbornadiene than with norbornene and cyclohexene, and earlier it was noted6 that norbornene was considerably more reactive than cyclohexene. As there is no evidence for the formation of resonance-stabilized radicals of type I or I T I , the enhanced reactivity of the bicyclic diolefin (and olefin) can probably be best rationalized in terms of relative destabilizations of reactants, owing to the highly strained olefinic bicyclic compounds. Acknowledgments.-The authors are indebted to the Shell Development Co. for generous support of this work, for fellowship support (G.D.B.) and (J.A.R.), as well as for samples of norbornadiene and 1,2,3,4-tetrachloro-5,5-diethoxycyclopentacliene. U'e are further indebted to H. Bluestone, R. E. Lidoo, J. Hyman and S . B. Solowny'0 for helpful discussions and for disclosure of their results prior to publication. Experimental Addition of +-Thiocresol to Norbornadiene.--p-Thiocresol (?i.6 g., 0.34 mole) was added in small portions with stir-

ring to 32.2 g. (0.35 mole) of norbornadiene with cooling in :in ice-snlt-bath. The reaction becomes dangerously exothermic if tlie hrat generated is not dissipated by an effective cooling medium. T h e reaction mixture was stirred for '70 minutes after t h e final addition of p-thiocresol, and was then whjected to v:icuuin distillation. A fraction boiling from 105-122" (1 n i i i i . ' ~ IKLS collected, which weighed 67.2 vc liquid was redistilled for prrparation of an anThe sample boiled at 125-131' (1 mm.) ; ~ n dh:id n rcfractire index T Z * O . ~ D 1.5840. )r CIIHIGS:C , 7 i . 7 2 ; 13, 7.46. Found: triple.

Preparation of 3-Nortricyclyl p-Tolyl Sulfone (X) .--I slight molar exceqs of 30y0hydrogen peroxide was added to a solution of 1.064 g. (4.93 rnrnoles) of the addition product in ( 2 7 ) Simonetla arid T\-insteiu?s have calculated t h e stabilization dne t o n-electron delocalization in t h e case of vinyl groups 6 t o a cationic carbon center. Their calcu1,itions indicate t h a t t h e nonclassical carbonium ion I X is considerably more stable t h a n the model electron-localized 5-norhornen-2-yl cation; this result is consistent with many experimental observations. If one utilizes t h e model a n d assimptian of Simonettu and Winstein for t h e cation and a d d s a third electron t o give t h e radical, the third electron will go into a non!>,indingorbitnl. but t!ie radicrrl will h a r e reSonance energy identical w i t h t h z l of t h e cation 21 If these calculations are proper a n d correct, t!ien t?ie two individual radicals m a y be best denoted by VI11 .rnd V I I . Erperiment:il evidence is n o t available a t present t o test I his posiihility, and we therefore will utilize t h e formulation of tlie sep:irate radicals as \'I and V I I . (28) h1. Simonetta and S. R'instein, THISJOURNAL, 76, 18 (1954). (20) This calculatinri W R L ~ n : d e h y n r . J~ deHeer. t o whom the

Feb. 5 , 1955

ADDITIONOF

THIOC CRESOL TO NORBORNADIENE

639

4 ml. of glacial acetic acid. The mixture was heated a t em-thioether XI1 described above, the melting point was 117.5-118.5'. reflux for one hour and enough additional hydrogen peroxide Diels-Alder Addition Product of 1,2,3,4-Tetrachloro-5,5mas added to give a strong test with starch-iodide paper. The mixture was heated a t reflux for two more hours; the diethoxycyclopentadiene and endo-5-Norbornen-2-y1 psolvent was then evaporated under reduced pressure. T h e Tolyl Thioether.-A solution of 1.26 g. (5.8 mmoles) of 5residue was extracted with two 75-1111. portions of low-boiling norbornen-2-yl p-tolyl thioether,6 which is known t o be petroleum ether. Concentration and cooling of the extract largely the endo isomer,6 and 2.56 g. (8.8 mmoles) of 1,2,3,4tetrachloro-5,5-diethoxycyclopentadienein 5 ml. of dry tolugave 392 mg. (327,) of white crystals, m.p. 69-76", Two recrystallizations from ethanol-water gave crystals of X melt- ene was heated a t reflux for 40 hours. After this time, the reaction mixture was steam distilled until the distillate ing a t 76-77' after drying. showed only traces of oil. Water was decanted from the Anal. Calcd. for CI4Hl7O2S:C, 67.71; H, 6.50. Found: gummy residue. This residue produced 1.0 g. (34%) of XV, C, 67.91; H , 6.54. m.p. 125-126". When mixed with the isomeric exo-thioThis compound did not decolorize dilute potassium per- ether X I I I , the m.p. was 104-108'. manganate in acetone or bromine in carbon tetrachloride. I t Anal. Calcd. for C2sH2&1402S: C, 54.34; H , 5.16. absorbed no hydrogen over a 10yo palladium-on-charcoal Found: C, 54.52; H , 5.05. catalyst. Preparation of 1,2,3,4-Tetrachloro-9,9-diethoxy-l,4,4a,Preparation of 1,2,3,4-Tetrachloro-9,9-diethoxy-l,4,4a,5 ,- 5,6,7,8,8a-octahydro - endo - p - toluenesulfonyl- 1,4,5,8 di6,7,8,8a-octahydro-7-exo-p- thiocresoxy- 1,4,5,8 - dimethano- methanonaphthalene- 7(XVII).-A mixture of 157 mg. (0.309 naphthalene (XII).-Al sample of the purified addition prod- mmole) of XVI, 5 ml. of pyridine and a slight excess of 30% uct weighing 1.022 g. (4.73 mmoles) was heated a t 150" with hydrogen peroxide was heated for 8.5 hr. on a steam-bath. 1.039 g. (3.56 mmoles) of 1,2,3,4-tetrachloro-5,5-diethoxy- T h e solvent was evaporated, and the residue was crystalcyclopentadiene for 4 hours. T h e reaction mixture was lized from low-boiling petroleum ether t o give 78 mg. (477,) chromatographed on a column of alumina. Elution with of a solid, m.p. 198-198.5'. Two recrystallizations from 1500 ml. of petroleum ether in 100-nil. fractions gave an oil petroleum ether gave XVII, m.p. 200.5-201'. which was probably diene. Elution with 200 ml. of carbon Anal. Calcd. for C23H2&la04S: C, 51.12; H , 4.85. tetrachloride in two portions gave 583 mg. (32%) of a solid, Found: C, 51.16; H , 4.92. m.p. 100-106". This solid was recrystallized twice from Addition of Thiophenol to Norbomadiene .-In a 200-1111. 9570 ethanol and gave crystals melting a t 117-117.5'. A three-necked flask equipped with a stirrer, thermometer and third ethanol recrystallization gave pure XI1 of 118-118.5'. arefluxcondenser, wasplaced56.1 g. (0.61 mole) of norbomaAnal. Calcd. for C23H20C1402S: C, 54.34; H, 5.16. diene. The diene was freshly distilled (b.p. 82-83.5' a t 627 Found: C, 54.43; H, 5.37. mm.) but was not kept under nitrogen. T o the stirred diene was added 62.2 ml. (67 g., 0.61 mole) of thiophenol from an Preparation of 1,2,3,4-Tetrachloro-9,9-diethoxy-l,2,3,4,4a,5,6,7,8,8a-octahydro-7-exo-ptoluenesulfonyl- 1,4,5,8-di- addition funnel through the reflux condenser. The mercapmethanonaphthalene (XIII).-A mixture of 116 mg. (0.228 tan was added t o the diene very slowly (several drops a t one mmole) of the thioether prepared in the experiment above, 5 time) a t the beginning of the reaction. It was added in ml. of pyridine and a small excess of 30q0 hydrogen peroxide larger portions as the reaction proceeded. The temperawas heated for 8 hr. on a steam-bath with addition of several ture of the reaction mixture was kept under 30" in about the drops of 307, hydrogen peroxide. T h e addition of water t o first third of the reaction, under 40" in the second third and the reaction mixture precipitated a fine oil which could be under 60" in the final third. The mercaptan was added over a 2.5-hr. period. coagulated by adding sodium chloride. The coagulated maThe colorless reaction mixture was subjected t o vacuum terial was crystallized from lor-boiling petroleum ether t o give a 647, yield of crystals, m.p. 216-222'. Recrystalliza- distillation and the fraction boiling a t 108-114" (0.8 mm.), tion from low-boiling petroleum ether followed by recrystal- NZ1D 1.5937 was collected. It consisted of 108.9 g. (88.67,) of a clear, straw-colored liquid. A hydrogenation experilization from 957, ethanol gave X I I I , m.p. 225-226O. ment30 indicated that this material contained about 407, Anal. Calcd. for C2aH2sC1404S:C, 51.12; H , 4.85. olefinic material. Found: C, 51.31; H, 5.03. Anal. Calcd. for C13H14S:C, 77.18; H, 6.98. Found: Formation of the Diels-Alder Addition Product of exo- C, 77.33; H , 7.13. 5-Norbornen-2-yl $-Tolyl Sulfone (XI) and 1,2,3,4-TetraPreparation of Nortricyclyl Phenyl Sulfone .--A mixture of chloro-S,S-diethoxycyclopentadiene.--,4 sample of 51 mg. 893 mg. (4.42 mmoles) of the norbomadiene-thiophenol (0.205 mmole) of exo-norbornen-2-yl p-tolyl sulfone6 was addition mixture, 5 ml. of dry acetic acid (containing acetic dissolved in 2 ml. of 1,2,3,4-tetrachloro-5,5-diethoxycyclo- anhydride) and 1.7 ml. of 30Y0 hydrogen peroxide was pentadiene and heated a t 138" for 12 hours. The reaction on a steam-bath for 30 minutes and allowed t o stand mixture was steam distilled until the distillate showed only aheated t room temperature for about 36 hours. Addition of ice traces of oil. \Vatu was decanted from the gummy residue and water precipitated a milky oil which later crystallized. which was then crystallized from 957, ethanol. The crys- The crystals were filtered and washed with 50 ml. of cold watals were washed with a little ethanol. The solid weighed ter. When dry they weighed 0.88 g. (85yc) and had a 81 mg. (737,), m.p. 220-221'. Another ethanol crystalliza- melting range of 57-77'. This material did not decolorize tion gave slightly impure X I I I , m.p. 221-222;. When mixed with the sulfone XI11 from the preceding experiment, dilute potassium permanganate in acetone. One recrystallization from petroleum ether gave 575 mg. (557,) of white it melted a t 221-223.5". crystals, m.p. 77-79'. Two additional recrystallizations Addition of #-Thiocresol to 1,2,3,4-Tetrachloro-9,9-dieth-from the same solvent gave needles, m.p. 79.5-80.1'. Thii oxy-l,4,4a,5,8.8a-hexahydro-l,4,5,8 - dimethanonaphthalene material not decolorize dilute potassium permanganate (XIV).--,4 sample of 3.5 g. (0 031 mole) of norbornadiene in acetonedid on standing several hours a t room temperature. was heated for 5 hours with 6.08 g. (0.026 mole) of 1,2,3,4- The infrared spectrum of this compound showed a strong abtetrachloro-5,5-dietho~yclopent3dientadiene in a sealed glass sorption peak a t 12.34 p . tube. A solid crystallized in the tube on standing a t room Anal. Calcd. for ClaHlaSOr:C, 66.61;H, 6.02. Found: temperature. Recrystallization from ethanol-water gave C, 66 81; H, 6.09. 2.98 g. (38Yc)of crystals which softened a t 80' and melted completely a t 105". Recrystallization from ethanol-water Hydrogenation of endo-5-Norbornen-2-y1 p-Tolyl Thiogave a solid, m.p. 93-94.5". X second crop from the first ether (XV) .-A sample of the endo-norbomenyl p-tolyl thiorecrystallization weighing 0.32 g. melted a t 92-93'. T h e ether weighing 436 mg. (2.02 mmoles) was hydrogenated in combined yield of impure X I V was 42%. .4 mixture of 214 4.5 ml. of ethanol with reduced 107, palladium-on-charcoal mg. (0.558 mmole) of XIV, 5 ml. of benzene, 75 mg. (0.6 catalyst. It absorbed 108 mole 7, of hydrogen (64.5 ml. a t mmole) of p-thiocresol and a trace of bicycloheptadiene (as a 26" and 630 mm.). source of peroxides) was heated a t reflux for one hour. T h e Competition of Norbornadiene and Norbornene for p solvent was then removed by evaporation. The addition of Thiocresol.-To a mixture of 921 mg. (10 mmoles) of nora few ml. of methanol t o the residue allowed a white solid t o bornadiene and 942 mg. (10 mmoles) of norbomene cooled in crystallize. The solid, weighing 111 mg. (40'",), m:p. an ice-bath, 1.304 g. (10.5 mmoles) of p-thiocresol was added 108-110", was recrystallized from a methanol-ethanol mixture to give XTI, m p. llfi-118°. W h e n miuerl with the (30) We are indebted t o Dr. R Heitz for this experiment.

-

640

K. D. HARDY A N D R. J. LTICKER

and the reaction allowed t o take place slowly. When the reaction was complete, the mixture was distilled a t atmospheric pressure (630 mm.). hlaterial boiling near 75" solidified in the condenser, and the plug of solid was pushed through into the receiver, which was cooled in a Dry Icebath. The weight of solid distillate was 0.63 g. T h e recovery of norbornene was EDYO. Competition of Norbornadiene and Cyclohexene for pThiocresol.-This experiment was carried out in essentially the same manner as that described above. T o a mixture of 939 mg. (10.2 mmoles) of norbornadiene and 822 mg. (10.0 mmoles) of cyclohexene a t Dry Ice temperature, 1.304 g. (10.5 mmoles) of p-thiocresol was added and the reaction allowed to take place as the reactants warmed t o room ternperature. The reaction mixture was distilled a t 630 mni., and 591 mg. (747,) of a liquid, b.p. 69', z Z o1.4465, ~ was collected. The refractive index of the starting cyclohexenr, %*OD, was 1.4465. Dilution Experiments with the Addition of p-Thiocresol to 4.49 Norbornadiene. First Exueriment.-Xorbornadiene. g. (48.6 mmoles) was stirr